J Neurol
DOI 10.1007/s00415-011-6272-x
ORIGINAL COMMUNICATION
Double-blind randomized trial on short-term efficacy
of the Semont maneuver for the treatment of posterior canal
benign paroxysmal positional vertigo
Marco Mandalà • Giovanni Paolo Santoro • Giacinto Asprella Libonati • Augusto Pietro Casani
Mario Faralli • Beatrice Giannoni • Mauro Gufoni • Vincenzo Marcelli • Pierpaolo Marchetti •
Emanuela Pepponi • Paolo Vannucchi • Daniele Nuti
•
Received: 20 August 2011 / Accepted: 29 September 2011
Ó Springer-Verlag 2011
Abstract The need for Class I and II studies on the
efficacy of Semont’s liberatory maneuver (SLM) in the
treatment of posterior canal benign paroxysmal positional
vertigo (PC-BPPV) motivated the present double-blind
randomized trial on the short-term efficacy of SLM. A total
of 342 patients with unilateral PC-BPPV were recruited for
a multicenter study. Patients were randomly assigned to
M. Mandalà (&) G. P. Santoro E. Pepponi D. Nuti
Dipartimento di Scienze Neurologiche e Sensoriali,
Università di Siena, Siena Viale Bracci, 11, 53100 Siena, Italy
e-mail: [email protected]
G. Asprella Libonati
Dipartimento di Otorinolaringoiatria,
Ospedale ‘‘Madonna delle Grazie, Matera, Italy
A. P. Casani
Dipartimento di Neuroscienze,
Università di Pisa, Pisa, Italy
M. Faralli
Dipartimento di Otorinolaringoiatria, Università di Perugia,
Perugia, Italy
B. Giannoni P. Vannucchi
Dipartimento di Scienze
Chirurgiche Oto-Neuro-Oftalmologiche,
Università di Firenze, Florence, Italy
M. Gufoni
Ambulatorio Otorinolaringoiatria, Azienda Ospedaliera 6,
Leghorn, Italy
V. Marcelli
Unità Operativa di Audiologia, Dipartimento di Neuroscienze,
Università di Napoli ‘Federico II’, Naples, Italy
P. Marchetti
Dipartimento di Statistica Medica ed Epidemiologia,
Università di Verona, Verona, Italy
treatment by SLM (n = 174) or sham treatment (n = 168).
Subjects were followed up twice (1 and 24 h) with the
Dix–Hallpike maneuver by blinded examiners. At the 1 and
24 h follow-up, 79.3 and 86.8%, respectively, of patients
undergoing SLM had recovered from vertigo, compared to
none of the patients undergoing the sham maneuver
(p \ 0.0001). Patients who manifested liberatory nystagmus at the end of SLM showed a significantly higher
percentage of recovery (87.1 vs. 55.7%; p \ 0.0001). To
the best of our knowledge, this is the first Class I study on
the efficacy of SLM. SLM proved highly effective with
respect to the sham maneuver (p \ 0.0001). Liberatory
nystagmus was demonstrated to be a useful prognostic
factor for the efficacy of treatment. The present Class I
study of efficacy of SLM changes the level of recommendation of the maneuver for treating PC-BPPV from
level C to level B.
Keywords Benign paroxysmal positional vertigo Semont’s liberatory maneuver Evidence-based medicine Double-blind randomized trial Vestibular Neuro-otology Semicircular canals
Introduction
Benign paroxysmal positional vertigo (BPPV) is a labyrinthine disorder caused by dislodged otoconia floating in
the semicircular canals (canalolithiasis) or, less frequently,
by otoconia attached to the cupula (cupulolithiasis).
Because of its anatomical position, the posterior canal (PC)
is the canal most frequently involved. The major treatment
for BPPV relies on physical maneuvers that enable the
otoconia to leave the canal by gravitation and centrifugal
inertia. Epley’s canalith repositioning procedure (CRP) and
123
J Neurol
Semont’s liberatory maneuver (SLM) are the most widely
used procedures for treating PC-BPPV. CRP and SLM
were first described in 1979 and 1983, respectively, and
subsequently simplified; the first became more popular in
the United States and the second in Europe [1]. Although
both maneuvers are considered highly effective by many
experts, the guidelines for treatment of BPPV produced by
the American Academy of Neurology [2], the American
Academy of Otolaryngology [3], and a recent review [4]
concluded that CRP is ‘an effective and safe treatment that
should be offered to patients of all ages’ with PC-BPPV
(Level A recommendation), while ‘SLM, based on currently published articles, can only be classified as ‘‘possibly
effective’’ (Level C recommendation) since Class I and II
studies are missing’.
The need to determine the evidence-based efficacy of
SLM, as highlighted by the ‘BPPV guidelines’ produced by
the American Academies of Neurology and Otolaryngology [2, 3], motivated the present trial. The main aim of the
study was to determine the short-term efficacy of SLM by a
double-blind randomized trial. The prognostic value of liberatory nystagmus (at the end of SLM) and short-term
follow-up (within 1 h) were also assessed.
Methods
A total of 342 patients with unilateral PC-BPPV were
recruited for a multicenter study that involved 7 otoneurology units in Italy (Florence, Leghorn, Matera, Naples,
Perugia, Pisa, and Siena) from March 2009 to December
2010. PC-BPPV was diagnosed according to the following
criteria [3]: (a) history of vertigo associated with changes
in head position; (b) torsional–vertical nystagmus (with the
upper pole of the eye beating toward the affected ear)
detected with Frenzel glasses or videoculography in the
Dix–Hallpike position; (c) vertigo associated with the
elicited nystagmus; (d) latency between completion of the
Dix–Hallpike test and the resulting vertigo and nystagmus
that increases and resolves within 1 min.
Patients with bilateral PC-BPPV, multiple canal atypical
positional nystagmus or who were treated previously with
repositioning maneuvers were excluded.
Patients were randomly assigned by a computer generated code to treatment by SLM (n = 174) or sham treatment (n = 168). Once the pathological side had been
identified by the Dix–Hallpike test, SLM was performed
with the examiner standing in front of the patient who was
seated on one side of the examining table with his/her legs
hanging freely. The patient’s head was rotated 45° to the
unaffected side, and then in a quick and continuous
movement, the patient was moved so as to lie on his
pathological side, with the back of the head resting on the
123
table. The patient was kept in this position for 2 min and
was then quickly brought up to the sitting position and then
lowered onto the opposite side, maintaining the head in the
same position relative to the shoulders. Care was taken not
to exceed 1.5 s in executing the 180° swing, to elicit sufficient acceleration on the canal to allow the debris to fall
into the utricle [5]. At the end of the maneuver, the patient
was lying on the shoulder of the unaffected side with the
cheekbone and nose in contact with the bed. In this position, the investigator checked for liberatory nystagmus
(nystagmus with the same direction as that evoked by the
Dix–Hallpike test) [6]. The patient was finally brought
back to the sitting position. The sham maneuver consisted
of SLM performed for the unaffected side. The maneuvers
were only performed once and patients did not receive any
subsequent instructions.
Then, 1 h and 24 h after SLM or the sham maneuver,
outcome was assessed by a second and a third blinded
investigator who repeated the Dix–Hallpike test without
interviewing the patient. Patients responding negatively
were considered to have recovered. Patients in both groups
who showed typical PC-BPPV at the 24 h follow-up were
treated again by SLM for the affected side.
Patients were instructed to return to the hospital if BPPV
symptoms persisted or reoccurred. Follow-ups of patients
recruited for the study were recorded up to 2 weeks after
treatment. Patients who still manifested PC-BPPV were
treated again by SLM.
The study was approved by the local Ethics Committees
of all hospitals where the clinical trial was conducted.
Informed consent was obtained from all participants.
Comparisons between groups were assessed by the Fisher’s
exact test or t-test, as appropriate, at a significance level of
p \ 0.05. Statistical analysis was performed with SPSS
software (SPSS, Inc., Chicago, IL, USA).
Results
Clinical and demographic data for the two populations
(SLM and sham groups) are shown in Table 1. Groups
did not differ in demographic or clinical baseline characteristics, only the affected side showed a prevalence of
the right canal in the SLM group but this was not statistically significant (p = 0.0508). All patients underwent
follow-up for up to 24 h after treatment. At the 1 h follow-up, 79.3% of patients who underwent SLM had
recovered, and 86.8% of subjects in the SLM group were
free of BPPV 24 h after treatment, whereas none of the
patients who underwent the sham maneuver recovered
(p \ 0.0001). No statistically significant differences were
observed between the 1 and 24 h follow-ups in the SLM
group (p = 0.0859).
J Neurol
Table 1 Clinical and
demographical characteristics of
the SLM and sham groups
Statistical analysis: * t test;
1
Fisher’s exact test
Semont
maneuver
(n = 174)
Sham
maneuver
(n = 168)
p value
Age
62.1 ± 15.1
63.9 ± 16.2
0.2884*
Sex (M/F)
69/105
51/117
0.09141
Affected side (R/L)
103/71
81/87
0.05081
Onset of PPV (days)
14.9 ± 23.5
21.2 ± 39.4
0.0723*
Previous PPV episode (y/n)
9/165
6/162
0.59991
Liberatory nystagmus at the end of the first maneuver (y/n)
142/32
0/168
\0.00011
1 h control (absence/presence of positional nystagmus)
24 h control (absence/presence of positional nystagmus)
138/36
151/23
0/168
0/168
\0.00011
\0.00011
1 h control after SLM in the sham group (absence/presence
of positional nystagmus)
–
135/33
In three subjects, positional nystagmus changed to typical lateral canal BPPV after SLM.
Patients who manifested liberatory nystagmus after
SLM for the affected side showed a significantly higher
percentage of recovery at the 1 h follow-up (87.1 vs.
55.7%; p \ 0.0001), regardless of whether SLM was performed as first treatment or 24 h after the sham maneuver
(Fig. 1). SLM was performed without serious adverse
effects in all patients. Side effects of SLM were transient
nausea, vomiting, and loss of balance, and occurred in 35
subjects.
Fig. 1 Outcomes at 1 h follow-up in relation to occurrence of
liberatory nystagmus at the end of the maneuver in SLM-treated and
sham-maneuver-treated subjects. The sham maneuver group received
SLM 24 h after the sham maneuver. Statistical analysis: Fisher’s
exact test
–
Discussion
In 1983, at the NES conference in Leuven, Alain Semont
presented a new therapeutic technique for PC-BPPV; the
technique subsequently became very popular in Europe.
The physical maneuver ‘would free the cupola using the
addition of the pressure of the endolymph and the inertia of
the heavy materials’. The results were quite extraordinary
but were not published until some time later [1]. Since
then, SLM has been simplified and a plausible explanation
provided for its peculiar mechanism of action. Despite its
widespread adoption by otoneurologists, SLM is only
considered ‘possibly effective’, due to a lack of Class I and
II studies.
To the best of our knowledge, this is the first Class I
study on the efficacy of SLM in the treatment of PC-BPPV.
SLM proved highly effective in the treatment of PC-BPPV
compared to the sham maneuver (p \ 0.0001). As postulated by other authors [7], to minimize the confounding
effect of spontaneous remission and highlight the effect of
the maneuver, we decided to perform a short-term followup. Despite the well-known phenomenon of fatigue that
can mimic successful treatment due to dispersion of particles in the canal after repeated positional maneuvers [7],
we did not find any statistically significant differences
between the 1 and the 24 h follow-ups. We confirmed the
utility of the so-called ‘liberatory nystagmus’ [6, 8] as a
prognostic factor for efficacy of treatment in most cases;
nevertheless, its absence does not necessarily mean that the
maneuver will be unsuccessful [8]. More than one
maneuver was needed to treat the PC-BPPV in 15% of
subjects.
A Class II study on SLM outcomes showed a significant
decrease in vertigo intensity and frequency after treatment
[9]. Another Class III study obtained a cure rate of 94.2%
with one maneuver [10], very close to the 86.8% recovery
123
J Neurol
achieved in the present trial. Both of these studies confirmed the efficacy of the SLM with respect to the sham
group.
Two Class I studies on CRP demonstrated a *90%
recovery rate of PC-BPPV after a few weeks, which is very
close to the present 24 h recovery rate [7, 11]. When
comparing the short-term (24 h) results after one maneuver
of either procedure (SLM or CRP), outcomes of SLM
(86.8% recovery) were slightly better than those of CRP
(80% recovery [7]). Unfortunately, there is currently
insufficient data to establish the relative efficacy of SLM
and CRP [2].
In conclusion, the present Class I study of efficacy of
SLM changes the level of recommendation of the method
for treating PC-BPPV from level C to level B. As in the
case of CRP, another Class I or II study will be necessary
to establish SLM as an effective and safe treatment recommended for subjects of all ages suffering from PCBPPV (level A recommendation).
Conflict of interest
3.
4.
5.
6.
7.
8.
None.
9.
References
1. Nuti D (2002) History of benign paroxysmal positional vertigo.
In: Guidetti G, Pagnini P (eds) Labyrintholithiasis-related paroxysmal positional vertigo. Excerpta medica. Elsevier, Amsterdam, pp 11–18
2. Fife TD, Iverson DJ, Lempert T, Furman JM, Baloh RW, Tusa
RJ, Hain TC, Herdman S, Morrow MJ, Gronseth GS, Quality
Standards Subcommittee, American Academy of Neurology
123
10.
11.
(2008) Practice parameter: therapies for benign paroxysmal
positional vertigo (an evidence-based review): report of the
Quality Standards Subcommittee of the American Academy of
Neurology. Neurology 70:2067–2074
Bhattacharyya N, Baugh RF, Orvidas L, Barrs D, Bronston LJ,
Cass S, Chalian AA, Desmond AL, Earll JM, Fife TD, Fuller DC,
Judge JO, Mann NR, Rosenfeld RM, Schuring LT, Steiner RW,
Whitney SL, Haidari J, American Academy of OtolaryngologyHead and Neck Surgery Foundation (2008) Clinical practice
guideline: benign paroxysmal positional vertigo. Otolaryngol
Head Neck Surg 139:S47–S81
Lee SH, Kim JS (2010) Benign paroxysmal positional vertigo.
J Clin Neurol 6:51–63
Faldon ME, Bronstein AM (2008) Head accelerations during
particle repositioning manoeuvres. Audiol Neurootol 13:345–356
Nuti D, Nati C, Passali D (2000) Treatment of benign paroxysmal
positional vertigo: no need for postmaneuver restrictions. Otolaryngol Head Neck Surg 122:440–444
von Brevern M, Seelig T, Radtke A, Tiel-Wilck K, Neuhauser H,
Lempert T (2006) Short-term efficacy of Epley’s manoeuvre: a
double-blind randomised trial. J Neurol Neurosurg Psychiatry
77:980–982
Soto-Varela A, Rossi-Izquierdo M, Santos-Pérez S (2011) Can
we predict the efficacy of the semont maneuver in the treatment
of benign paroxysmal positional vertigo of the posterior semicircular canal? Otol Neurotol 32:1008–1011
Cohen HS, Kimball KT (2005) Effectiveness of treatments for
benign paroxysmal positional vertigo of the posterior canal. Otol
Neurotol 26:1034–1040
Salvinelli F, Trivelli M, Casale M, Firrisi L, Di Peco V,
D’Ascanio L, Greco F, Miele A, Petitti T, Bernabei R (2004)
Treatment of benign positional vertigo in the elderly: a randomized trial. Laryngoscope 114:827–831
Lynn S, Pool A, Rose D, Brey R, Suman V (1995) Randomized
trial of the canalith repositioning procedure. Otolaryngol Head
Neck Surg 113:712–720